Filtering comparisons for vst data

Coloured by Diagnosis:

ggplotly(pcas_samps %>% ggplot(aes(PC1, PC2, color=Diagnosis_)) + geom_point(aes(frame=lfc, ids=ID)) + 
         theme_minimal() + ggtitle('Samples PCA plot modifying filtering threshold'))

Coloured by brain region:

No recognisable pattern

ggplotly(pcas_samps %>% ggplot(aes(PC1, PC2, color=Brain_lobe)) + geom_point(aes(frame=lfc, ids=ID)) + 
         theme_minimal() + ggtitle('Samples PCA plot modifying filtering threshold'))

SFARI genes coloured by score

pcas_sfari_genes = pcas_genes %>% filter(!is.na(score)) %>% dplyr::select(-'score')

complete_sfari_df = expand.grid(unique(pcas_sfari_genes$ID), unique(pcas_sfari_genes$lfc))
colnames(complete_sfari_df) = c('ID', 'lfc')

pcas_sfari_genes = pcas_sfari_genes %>% right_join(complete_sfari_df, by=c('ID','lfc')) %>% 
                   left_join(SFARI_genes, by='ID') %>% 
                   mutate('score'=as.factor(`gene-score`), 'syndromic'=as.factor(syndromic))
pcas_sfari_genes[is.na(pcas_sfari_genes)] = 0 # Fix for ghost points
  
ggplotly(pcas_sfari_genes %>% ggplot(aes(PC1, PC2, color=score)) + 
         geom_point(aes(frame=lfc, ids=ID), alpha=0.6) + theme_minimal() + 
         ggtitle('Genes PCA plot modifying filtering threshold'))

Most of the genes get filtered out by the first adjusted p-value<0.05 filter (including all the genes with score=1 except two), but the proportion of genes left after the first cut is significantly higher for all scores and they seem to be generally filtered out after as well

table(SFARI_genes$`gene-score`[SFARI_genes$ID %in% genes_DE_info$ID[genes_DE_info$adj.P.Val<0.05]])

## 
##  1  2  3  4  5  6 
##  2  8 33 77 45  6

# Calculate percentage of genes remaining on each lfc by each score
score_count_by_lfc = pcas_genes %>% filter(!is.na(score)) %>% group_by(lfc, score) %>% tally %>% ungroup
score_count_pcnt = score_count_by_lfc %>% filter(lfc==-1) %>% mutate('n_init'=n) %>%
                   dplyr::select(score, n_init) %>% right_join(score_count_by_lfc, by='score') %>%
                   mutate('pcnt'=round(n/n_init*100, 2)) %>% filter(lfc!=-1)

# Complete missing entries with zeros
complete_score_count_pcnt = expand.grid(unique(score_count_pcnt$lfc), unique(score_count_pcnt$score))
colnames(complete_score_count_pcnt) = c('lfc', 'score')
score_count_pcnt = full_join(score_count_pcnt, complete_score_count_pcnt, by=c('lfc','score')) %>%
                   dplyr::select(score, lfc, n, pcnt)
score_count_pcnt[is.na(score_count_pcnt)] = 0

# Join counts by score and all genes
all_count_pcnt = pcas_genes %>% group_by(lfc) %>% tally  %>% filter(lfc!=-1) %>% 
                 mutate('pcnt'=round(n/nrow(datExpr)*100, 2), 'score'='All')
score_count_pcnt = rbind(score_count_pcnt, all_count_pcnt)

ggplotly(score_count_pcnt %>% ggplot(aes(lfc, pcnt, color=score)) + geom_point() + geom_line() + 
         scale_colour_manual(palette=gg_colour_hue) + theme_minimal() + 
         ggtitle('% of points left after each increase in log2 fold change'))

rm(score_count_by_lfc, complete_score_count_pcnt)

SFARI genes coloured by syndromic tag

Most syndromic genes get filtered out with the p-value threshold, the remaining ones don’t seem to have a very different behaviour to the rest of the data, perhaps they survive a bit longer.

ggplotly(pcas_sfari_genes %>% ggplot(aes(PC1, PC2, color=ordered(syndromic, levels=c(1,0)))) + 
         geom_point(aes(frame=lfc, ids=ID), alpha=0.6) + theme_minimal() + 
         scale_colour_manual(palette=gg_colour_hue) +
         ggtitle('Genes PCA plot modifying filtering threshold'))

# Calculate percentage of syndromic genes remaining on each lfc
syndromic_count_by_lfc = pcas_sfari_genes %>% filter(syndromic==1 & PC1!=0) %>% group_by(lfc) %>% tally %>% 
                         ungroup %>% filter(lfc!=-1) %>% 
                         mutate('pcnt' = round(n/nrow(SFARI_genes[SFARI_genes$syndromic==1,])*100,2), 'score'='syndromic')

# Complete missing entires with zeros and add stats for all genes for comparison
syndromic_count_by_lfc = data.frame('lfc' = unique(pcas_genes$lfc), 'score'='syndromic') %>% filter(lfc!=-1) %>% 
                         full_join(syndromic_count_by_lfc, by=c('lfc','score')) %>% replace(.,is.na(.),0) %>%
                         rbind(all_count_pcnt) %>% mutate('score'=ordered(score, levels=c('syndromic','All')))


ggplotly(syndromic_count_by_lfc %>% ggplot(aes(lfc, pcnt, color=score)) + geom_point() + geom_line() + 
         scale_colour_manual(palette=gg_colour_hue) + theme_minimal() + 
         ggtitle('% of points left after each increase in log2 fold change'))

All genes together

ggplotly(pcas_genes %>% ggplot(aes(PC1, PC2)) + geom_point(aes(frame=lfc, ids=ID, alpha=0.3)) + 
         theme_minimal() + ggtitle('Genes PCA plot modifying filtering threshold'))